I'm building a schedule from a partitioned job. I know that the partition value is passed to the job using

dagster/partition

in

tags

. But how can I make this value dynamic? I think incremental refresh is the most common use case of a fact table in data warehouse. I'm stuck on how to implement that. Can anyone help?

You have to use a dynamic partition. You can use a sensor to add partitions, I think you could also use an asset or an op, but I am not sure about it.

Hi @燕子石! If you want to kick off a run of a specific partition, you can do this within the body of an

@schedule

-decorated function:

@schedule(job=my_job, cron_schedule="0 0 * * *")
def configurable_job_schedule(context: ScheduleEvaluationContext):
    return RunRequest(
        run_key=None,
        partition_key="2023-08-01",
    )